Identify advantages of mixed methods research and describe specific applications

  Describe strategies and designs for conducting mixed methods research

  Identify the purposes and some of the distinguishing features of specific types of research (e.g., clinical trials, evaluations, outcomes research, surveys)

  Define new terms in the chapter

Key Terms

   Clinical trial

   Concurrent design

   Convergent design

   Delphi survey

   Economic (cost) analysis

   Evaluation research

   Explanatory design

   Exploratory design

   Health services research

   Intervention research

   Intervention theory

   Methodologic study

   Mixed methods research

   Nursing sensitive outcome

   Outcomes research

   Pragmatism

   Process analysis

   Quality improvement (QI)

   Secondary analysis

   Sequential design

   Surveys

In this final chapter on research designs, we explain several special types of research. We begin by discussing mixed methods research that combines quantitative and qualitative approaches.

MIXED METHODS RESEARCH

A growing trend in nursing research is the planned collection and integration of quantitative and qualitative data within a single study or coordinated clusters of studies. This section discusses the rationale for such mixed methods research and presents a few applications.

Rationale for Mixed Method Research

The dichotomy between quantitative and qualitative data represents a key methodologic distinction. Some argue that the paradigms that underpin quantitative and qualitative research are incompatible. Most people, however, now believe that many areas of inquiry can be enriched by triangulating quantitative and qualitative data. The advantages of a mixed methods (MM) design include the following:

  Complementarity. Quantitative and qualitative approaches are complementary. By using mixed methods, researchers can possibly avoid the limitations of a single approach.

  Practicality. Given the complexity of phenomena, it is practical to use whatever methodological tools are best suited to addressing pressing research questions.

  Enhanced validity. When a hypothesis or model is supported by multiple and complementary types of data, researchers can be more confident about their inferences.

Perhaps the strongest argument for MM research, however, is that some questions require MM. Pragmatism, a paradigm often associated with MM research, provides a basis for a position that has been stated as the “dictatorship of the research question” (Tashakkori & Teddlie, 2003, p. 21). Pragmatist researchers consider that it is the research question that should drive the design of the inquiry. They reject a forced choice between the traditional postpositivist and constructivist approaches to research.

Purposes and Applications of Mixed Methods Research

In MM research, there is typically an overarching goal, but there are inevitably at least two research questions, each of which requires a different type of approach. For example, MM researchers may simultaneously ask exploratory (qualitative) questions and confirmatory (quantitative) questions. In an MM study, researchers can examine causal effects in a quantitative component but can shed light on causal mechanisms in a qualitative component.

Creswell and Plano Clark (2011) identified six types of research situations that are especially well suited to MM research:

1.  The concepts are new and poorly understood, and there is a need for qualitative exploration before more formal, structured methods can be used.

2.  Neither a qualitative nor a quantitative approach, by itself, is adequate in addressing the complexity of the research problem.

3.  The findings from one approach can be greatly enhanced with a second source of data.

4.  The quantitative results are puzzling and difficult to interpret, and qualitative data can help to explain the results.

5.  A particular theoretical perspective might require both quantitative and qualitative data.

6.  A multiphase project is needed to attain key objectives, such as the development and assessment of an intervention.

As this list suggests, MM research can be used in various situations. Some of the major applications include the following:

  Instrument development. Nurse researchers sometimes gather qualitative data as the basis for developing formal instruments—that is, for generating and wording the questions on quantitative scales that are subsequently subjected to rigorous testing.

  Intervention development. Qualitative research is also playing an important role in the development of promising nursing interventions that are then rigorously tested for efficacy.

  Hypothesis generation. In-depth qualitative studies are often fertile with insights about constructs or relationships among them. These insights then can be tested and confirmed with larger samples in quantitative studies.

  Theory building and testing. A theory gains acceptance as it escapes disconfirmation, and the use of multiple methods provides opportunity for potential disconfirmation of a theory. If the theory can survive these assaults, it can provide a stronger context for the organization of clinical and intellectual work.

  Explication. Qualitative data are sometimes used to explicate the meaning of quantitative descriptions or relationships. Quantitative methods can demonstrate that variables are systematically related but may fail to explain why they are related.

Mixed Method Designs and Strategies

In designing MM studies, researchers make many important decisions. We briefly describe a few.

Design Decisions and Notation

Two decisions in MM design concern sequencing and prioritization. There are three options for sequencing components of an MM study: Qualitative data are collected first, quantitative data are collected first, or both types are collected simultaneously. When the data are collected at the same time, the approach is concurrent. The design is sequential when the two types of data are collected in phases. In well-conceived sequential designs, the analysis and interpretation in one phase informs the collection of data in the second.

In terms of prioritization, researchers usually decide which approach—quantitative or qualitative—to emphasize. One option is to give the two components (strands) equal, or roughly equal, weight. Usually, however, one approach is given priority. The distinction is sometimes referred to as equal status versus dominant status.

Janice Morse (1991), a prominent nurse researcher, made a major contribution to MM research by proposing a widely used notation system for sequencing and prioritization. In this system, priority is designated by uppercase and lowercase letters: QUAL/quan designates an MM study in which the dominant approach is qualitative, whereas QUAN/qual designates the reverse. If neither approach is dominant (i.e., both are equal), the notation is QUAL/QUAN. Sequencing is indicated by the symbols + or →. The arrow designates a sequential approach. For example, QUAN → qual is the notation for a primarily quantitative MM study in which qualitative data are collected in phase 2. When both approaches occur concurrently, a plus sign is used (e.g., QUAL + quan).

Specific Mixed Methods Designs

Numerous design typologies have been proposed by different MM methodologists. We illustrate a few basic designs described by Creswell (2015).

The purpose of the convergent design (sometimes called a triangulation design) is to obtain different, but complementary, data about the central phenomenon under study—i.e., to triangulate data sources. The goal of this design is to converge on “the truth” about a problem or phenomenon by allowing the limitations of one approach be offset by the strengths of the other. In this design, quantitative and qualitative data are collected simultaneously, with equal priority (QUAL + QUAN).

Explanatory designs are sequential designs with quantitative data collected in the first phase, followed by qualitative data collected in the second phase. Either the quantitative or the qualitative strand can be given a stronger priority: The design can be either QUAN → qual or quan → QUAL. In explanatory designs, qualitative data from the second phase are used to build on or explain the quantitative data from the initial phase. This design is especially suitable when results are complex and tricky to interpret.

Exploratory designs are sequential MM designs, with qualitative data being collected first. The design has as its central premise the need for initial in-depth exploration of a concept. Usually, the first phase focuses on exploration of a poorly understood phenomenon, and the second phase is focused on measuring it or classifying it. In an exploratory design, either the qualitative phase can be dominant (QUAL → quan) or the quantitative phase can be dominant (qual → QUAN).

TIP Creswell and Plano Clark (2011) described a design called the embedded design—a term that is sometimes used in nursing studies. However, Creswell (2015) subsequently stopped referencing this design. An embedded design is one in which a second type of data is totally subservient to the other type of data. Creswell now views embedding as an analytic strategy rather than as a design type.

Sampling and Data Collection in Mixed Methods Research

Sampling and data collection in MM studies are often a blend of approaches described in earlier chapters. A few special issues for an MM study merit brief discussion.

MM researchers can combine sampling designs in various ways. The quantitative component is likely to rely on a sampling strategy that enhances the researcher’s ability to generalize from the sample to a population. For the qualitative component, MM researchers usually adopt purposive sampling methods to select information-rich cases who are good informants about the phenomenon of interest. Sample sizes are also likely to be different in the quantitative and qualitative strands in ways one might expect—i.e., larger samples for the quantitative component. A unique sampling issue in MM studies concerns whether the same people will be in both the quantitative and qualitative strands. The best strategy depends on the study purpose and the research design, but using overlapping samples can be advantageous. Indeed, a particularly popular strategy is a nested approach in which a subset of participants from the quantitative strand is used in the qualitative strand.

In terms of data collection, all of the data collection methods discussed previously can be creatively combined and triangulated in an MM study. Thus, possible sources of data include group and individual interviews, psychosocial scales, observations, biophysiological measures, records, diaries, and so on. MM studies can involve intramethod mixing (e.g., structured and unstructured self-reports) and intermethod mixing (e.g., biophysiologic measures and unstructured observation). A fundamental issue concerns the methods’ complementarity—that is, having the limitations of one method be balanced and offset by the strengths of the other.

TIP One challenge in doing MM research concerns how best to analyze the quantitative and qualitative data. The benefits of MM research require an effort to merge results from the two strands and to develop interpretations and recommendations based on integrated understandings.

OTHER SPECIAL TYPES OF RESEARCH

The remainder of this chapter briefly describes types of research that vary by study purpose rather than by research design or tradition.

Intervention Research

In Chapter 9, we discussed randomized controlled trials (RCTs) and other experimental and quasi-experimental designs for testing the effects of interventions. In actuality, intervention research is often more complex than a simple experimental–control group comparison of outcomes—indeed, intervention research often relies on MM to develop, refine, test, and understand the intervention.

Different disciplines have developed their own approaches and terminology in connection with intervention efforts. Clinical trials are associated with medical research, evaluation research is linked to the fields of education and public policy, and nurses are developing their own tradition of intervention research. We briefly describe these three approaches.

Clinical Trials

Clinical trials test clinical interventions. Clinical trials undertaken to evaluate an innovative therapy or drug are often designed in a series of phases:

  Phase I of the trial is designed to establish safety, tolerance, and dose with a simple design (e.g., one-group pretest–posttest). The focus is on developing the best treatment.

  Phase II is a pilot test of treatment effectiveness. Researchers see if the intervention is feasible and acceptable and holds promise. This phase is designed as a small-scale experiment or a quasi-experiment.

  Phase III is a full experimental test of the intervention—an RCT with random assignment to treatment conditions. The objective is to develop evidence about the treatment’s efficacy—i.e., whether the intervention is more efficacious than usual care or another alternative. When the term clinical trial is used, it often is referring to a phase III trial.

  Phase IV of clinical trials involves studies of the effectiveness of an intervention in the general population. The emphasis in effectiveness studies is on the external validity of an intervention that has demonstrated efficacy under controlled (but artificial) conditions.

Evaluation Research

Evaluation research focuses on developing useful information about a program or policy—information that decision makers need on whether to adopt, modify, or abandon the program.

Evaluations are undertaken to answer various questions. Questions about program effectiveness rely on experimental or quasi-experimental designs, but other questions do not. Many evaluations are MM studies with distinct components.

For example, a process analysis is often undertaken to obtain descriptive information about the process by which a program gets implemented and how it actually functions. A process analysis addresses such questions as the following: What exactly is the treatment, and how does it differ from traditional practices? What are the barriers to successful program implementation? How do staff and clients feel about the intervention? Qualitative data play a big role in process analyses.

Evaluations may also include an economic (or cost) analysis to assess whether program benefits outweigh its monetary costs. Administrators make decisions about resource allocation for health services not only on the basis of whether something “works” but also based on economic viability. Cost analyses are often done when researchers are also evaluating program efficacy.

Stay updated, free articles. Join our Telegram channel

Join